The dorsal cochlear nucleus (DCN) integrates diverse multisensory inputs with auditory nerve signals to compute the location of a sound relative to body position. Loss of auditory input to DCN contributes to enhancement of multisensory input and could lead to hyperactivity in DCN, an event highly associated with tinnitus. Elucidating the multisensory circuitry of DCN is critical for both understanding auditory processing and pathophysiology that occurs in tinnitus. Multisensory input from disparate brain regions is carried to DCN by mossy fibers (MFs). MFs innervate granule cells and unipolar brush cells (UBCs). UBCs are glutamatergic interneurons that receive a single MF input at their brush-like dendrite and project to ensembles of granule cells whose parallel fiber axons modulate the activity of principal (fusiform) cells of the DCN. Recently, the Trussell lab has shown that UBCs respond to MF inputs by either markedly increasing or decreasing firing for seconds, depending on their ON or OFF subtype. ON UBCs could amplify the signal from a single MF input, synchronizing and enhancing the firing of numerous postsynaptic granule cells. Thus, in addition to amplifying specific multisensory channels of unknown origin, UBCs could contribute to pathologically enhanced DCN activity that is associated with tinnitus. To understand how multisensory integration occurs it is essential to identify the nature of the inputs to UBCs and how UBCs transmit information to other cells in the DCN circuit. The purpose of this proposal is to test: (1) where the neurons that project MFs to UBCs originate and what sensory modalities they represent, and (2) how the information they provide is integrated in the DCN circuit. In Aim 1 I will use cutting-edge anatomical tracing methods to identify MF projections that innervate UBCs. This will elucidate the source and sensory modality of the signals processed by UBCs. Several candidate regions that project MFs to DCN carry proprioceptive, motor and higher-level auditory feedback information, but it is unclear whether they innervate UBCs. I will Identify which sources innervate UBCs and whether both ON and OFF UBC subtypes are targeted. In Aim 2 I will characterize the function of identified MF input to UBCs electrophysiologically by expressing channelrhodopsin in UBC projecting sources. Using 2-photon microscopy I will define the spatial projection pattern of UBC axons within DCN and make paired electrophysiological recordings to test how MF input is transformed between UBCs and postsynaptic granule cells. This research will identify a major missing piece of the DCN circuit: what information UBCs process and their effect on the granule cell system. Transformation of this information by UBCs is likely to play a major role in multisensory integration and sound source localization in DCN. Because of UBCs' potential role in the amplification of excitatory signals, this work may provide insights into tinnitus, a common disorder associated with DCN hyperactivity.